http://www.k-wave.org/forum/topic/problems-about-gpu-memory-used Support for the k-Wave MATLAB toolbox en-US Tue, 12 May 2026 22:28:38 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/problems-about-gpu-memory-used#post-8758 Wed, 17 May 2023 18:53:42 +0000 Bradley Treeby 8758@http://www.k-wave.org/forum/ <p>I’d advise switching to the C++/CUDA version of the GPU code - it uses much less memory in general, and the memory usage can be calculated reasonably accurately. Unfortunately, when using gpuArray, I’m not exactly sure how the inbuilt memory management of MATLAB works under the hood, and why the memory usage might change in these two cases. </p> http://www.k-wave.org/forum/topic/problems-about-gpu-memory-used#post-8713 Tue, 07 Mar 2023 09:52:22 +0000 zhaoxiang 8713@http://www.k-wave.org/forum/ <p>Hi, firstly I have to say this toolbox is very useful. However, I have a problem that has troubled me a lot.</p> <p>The problem is about GPU memory used. The code is run on A100(40GB). It looked normal at first：</p> <p>estimated simulation time 2hours 6min 29.7093s...<br /> GPU memory used: 21.9727 GB (of 39.5871 GB)<br /> simulation completed in 2hours 6min 36.3867s</p> <p>It looked like it was over but next something happende：</p> <p>{Error using gpuArray/ifft<br /> Out of memory on device. MATLAB was unable to allocate sufficient resources on<br /> the GPU to complete this operation. If the problem persists, reset the GPU by<br /> calling 'gpuDevice(1)'.</p> <p>Error in fourierShift (line 112)<br /> data = real(ifft(bsxfun(@times, ifftshift( exp(1i .* k_vec .* shift) ),<br /> fft(data, [], dim)), [], dim));</p> <p>Error in kspaceFirstOrder_saveIntensity (line 46)<br /> ux_sgt = fourierShift(sensor_data(sensor_mask_index).ux_non_staggered,<br /> 1/2);</p> <p>Error in kspaceFirstOrderAS (line 1387)<br /> kspaceFirstOrder_saveIntensity;</p> <p>Error in code1 (line 169)<br /> sensor_data = kspaceFirstOrderAS(kgrid, medium, source, sensor, ...<br /> } </p> <p>And I have another question about GPU memory used.<br /> There is another case which can be run on A100 and the GPU memory used is 3.5964 GB (of 39.5871 GB). However, this case cannot not be run on a GPU of 6GB.</p> <p>Could anybody tell me why?<br /> Thanks in advance. </p>